初始化项目,由ModelHub XC社区提供模型
Model: Bykot/rugpt3small_isolation_forest Source: Original Platform
This commit is contained in:
2
results/cleaning_stats.csv
Normal file
2
results/cleaning_stats.csv
Normal file
@@ -0,0 +1,2 @@
|
||||
experiment,method,source_docs,cleaned_docs,removed_docs,removed_percent,perplexity_model_name,contamination,n_estimators,max_samples,bootstrap,score_min,score_q05,score_q50,score_q95,score_max,created_at_utc
|
||||
isolation_forest,isolation_forest,198000,178200,19800,10.0,Qwen/Qwen2.5-0.5B,0.1,100,auto,False,-0.29584173481435616,-0.04286222651318166,0.11124878822941547,0.1459759723764503,0.1556654965682724,2026-05-04T21:44:08.636625+00:00
|
||||
|
11
results/cleaning_summary.json
Normal file
11
results/cleaning_summary.json
Normal file
@@ -0,0 +1,11 @@
|
||||
{
|
||||
"experiment": "isolation_forest",
|
||||
"method": "isolation_forest",
|
||||
"base_dataset_repo": "Bykot/c4_ru_200k_split",
|
||||
"source_train_docs": 198000,
|
||||
"cleaned_docs": 178200,
|
||||
"removed_docs": 19800,
|
||||
"removed_percent": 10.0,
|
||||
"eval_docs": 2000,
|
||||
"created_at_utc": "2026-05-04T21:44:09.794142+00:00"
|
||||
}
|
||||
3
results/features_analysis.csv
Normal file
3
results/features_analysis.csv
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be95987971eafeb5b86b70013d8e4c901b33ab58fb6ba553602d38479a378a5d
|
||||
size 27785771
|
||||
|
3
results/isolation_forest_model.pkl
Normal file
3
results/isolation_forest_model.pkl
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:9d10be87ad31fe4852c09e75a49a5da63f01dd5cacfbd6304c8b7cfaac7752f3
|
||||
size 836577
|
||||
2
results/isolation_forest_stats.csv
Normal file
2
results/isolation_forest_stats.csv
Normal file
@@ -0,0 +1,2 @@
|
||||
experiment,method,source_docs,cleaned_docs,removed_docs,removed_percent,perplexity_model_name,contamination,n_estimators,max_samples,bootstrap,score_min,score_q05,score_q50,score_q95,score_max,created_at_utc
|
||||
isolation_forest,isolation_forest,198000,178200,19800,10.0,Qwen/Qwen2.5-0.5B,0.1,100,auto,False,-0.29584173481435616,-0.04286222651318166,0.11124878822941547,0.1459759723764503,0.1556654965682724,2026-05-04T21:44:08.636625+00:00
|
||||
|
3
results/scaler.pkl
Normal file
3
results/scaler.pkl
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:80477fc868161103a173eb66aa55ba8deaa1d9109ec8e4641433b8b6865b0fd8
|
||||
size 871
|
||||
2
results/training_metrics.csv
Normal file
2
results/training_metrics.csv
Normal file
@@ -0,0 +1,2 @@
|
||||
experiment,method,model_name,base_dataset_repo,train_docs,removed_docs,eval_docs,eval_loss_before,perplexity_before,eval_loss_after,perplexity_after,train_runtime,train_samples_per_second,created_at_utc
|
||||
isolation_forest,isolation_forest,ai-forever/rugpt3small_based_on_gpt2,Bykot/c4_ru_200k_split,178200,19800,2000,3.8492491245269775,46.95779053735424,3.059091329574585,21.308186243810518,16459.4506,10.827,2026-05-05T02:25:19.582561+00:00
|
||||
|
16
results/training_metrics.json
Normal file
16
results/training_metrics.json
Normal file
@@ -0,0 +1,16 @@
|
||||
{
|
||||
"experiment": "isolation_forest",
|
||||
"method": "isolation_forest",
|
||||
"model_name": "ai-forever/rugpt3small_based_on_gpt2",
|
||||
"base_dataset_repo": "Bykot/c4_ru_200k_split",
|
||||
"train_docs": 178200,
|
||||
"removed_docs": 19800,
|
||||
"eval_docs": 2000,
|
||||
"eval_loss_before": 3.8492491245269775,
|
||||
"perplexity_before": 46.95779053735424,
|
||||
"eval_loss_after": 3.059091329574585,
|
||||
"perplexity_after": 21.308186243810518,
|
||||
"train_runtime": 16459.4506,
|
||||
"train_samples_per_second": 10.827,
|
||||
"created_at_utc": "2026-05-05T02:25:19.582561+00:00"
|
||||
}
|
||||
Reference in New Issue
Block a user